Articles · Great investors
Jim Simons and Renaissance Technologies: the Medallion Fund record
How mathematician Jim Simons built Renaissance Technologies, what is reported about the Medallion Fund, why it is closed and how the public funds fared.
On this page
Jim Simons (1938–2024) was a prize-winning mathematician who, in middle age, decided that markets could be treated as a data problem. The firm he founded, Renaissance Technologies, runs the Medallion Fund, whose reported results are the strongest of any investment fund on record: roughly 66% a year on average before fees and 39% after fees from 1988 to 2018, according to Gregory Zuckerman's 2019 history of the firm, with no losing year in that span.
Two facts make the story more instructive than a list of big numbers. First, almost nobody can invest in Medallion: since the early 2000s it has been run for the firm's own employees. Second, Renaissance also runs funds for outside investors, built by the same people, and those funds have had ordinary results, including large losses in 2020 while Medallion gained 76%. This article covers who Simons was, what is known about how the firm works, the record as reported, and what the gap between the two kinds of fund says about edge, capacity and luck.
From mathematics to markets
Simons was born in 1938 near Boston, studied mathematics at MIT and earned his doctorate at Berkeley in his early twenties. In the mid-1960s he worked as a codebreaker at the Institute for Defense Analyses, a contractor for the National Security Agency, until he was dismissed in 1968 after publicly opposing the Vietnam War. He then chaired the mathematics department at Stony Brook University. His work with Shiing-Shen Chern on what are now called Chern–Simons forms, published in 1974, later became important in theoretical physics, and in 1976 he won the American Mathematical Society's Oswald Veblen Prize in geometry.
In 1978 he left academia and started a trading firm, Monemetrics, renamed Renaissance Technologies in 1982. The early years were not quantitative in the modern sense. Simons and his partners traded currencies and commodities partly on judgment and had a rough ride. The shift came as they hired more mathematicians and computer scientists and moved to fully automated models. The Medallion Fund, named after the mathematics prizes several of its founders had won, began trading in 1988. It lost money in its first full year of the new approach, was rebuilt around shorter-term trading signals with help from the mathematician Elwyn Berlekamp, and gained sharply in 1990.
How Renaissance is described
The firm publishes almost nothing about its methods, so everything here comes from former employees, court and congressional records, and Zuckerman's reporting. Treat it as a broad outline, not a recipe.
- Data first. Renaissance collected and cleaned price data, going back decades and across many markets, long before that was common. Its researchers searched it for small, repeatable patterns, many of them tied to how prices move over hours and days.
- Many small edges. No single signal was strong. Zuckerman quotes the firm's people as saying they were right on only slightly more than half of their trades. The profits came from making a very large number of them, each with a small expected gain, and combining thousands of signals in one model.
- Leverage and hedging. Small edges become large returns only with leverage, which the fund used heavily, and with hedges that remove broad market exposure so that the net result depends on the signals, not on whether stocks rise.
- Scientists, not financiers. Simons hired mathematicians, physicists, astronomers and speech-recognition researchers, notably Robert Mercer and Peter Brown, who later ran the firm. They worked in one shared model rather than as competing teams.
- Short holding periods. Medallion's typical positions are reported to last from minutes to days. That matters for the next point.
Why Medallion is closed
Short-term patterns are limited in how much money they can absorb. Each trade moves prices a little, and the more money chasing the same signal, the less of the edge is left after trading costs. Renaissance responded by capping Medallion's size, reportedly around $10 billion, and paying out profits each year rather than letting the fund grow. It stopped taking new outside money in the 1990s and had returned outside investors' capital by around 2005. Today it is owned by current and former employees.
That is also why it can charge its reported fees of 5% of assets and 44% of profits (to insiders, who pay them to themselves in effect) and still report net returns of about 39% a year.
The record as reported
Three cautions apply to every Medallion number.
- Source. The figures come from fund records Zuckerman obtained, not from audited public filings. They have been widely repeated and never publicly disputed by the firm, but they are reported numbers.
- Averages are not compound rates. An average of yearly returns of 39% does not mean money compounded at 39% a year in practice, partly because Medallion paid profits out each year rather than reinvesting them. The average is the fairer way to describe a fund that does not compound its capital.
- Comparison. The S&P 500 bar is a different kind of number (price only, an index, not a leveraged hedged fund). It is there for scale, not as a like-for-like benchmark.
Individual years are just as unusual. Institutional Investor, citing past results, listed 2000 (98.5%) and 2008 (82.4%), figures first reported by the Wall Street Journal, as Medallion's only better years than 2020, when it gained 76%. Those three years were among the worst for most other investors, which suggests the fund's short-term signals did best when markets were most disorderly.
The public funds: same firm, ordinary results
In 2005 Renaissance launched the Renaissance Institutional Equities Fund (RIEF) for outside investors, and later the Renaissance Institutional Diversified Alpha (RIDA) fund and others. These funds are built by the same research team, but they are designed differently: they hold positions for much longer (months rather than days) so that they can manage tens of billions of dollars without moving prices. Longer horizons mean different, weaker and more crowded signals.
The contrast in 2020 was stark. Medallion gained 76%. RIEF lost 22.6% through 25 December by HSBC's tally (Bloomberg, citing investor letters, put the full-year loss at about 19%), and RIDA lost about 33.6% by HSBC's count. The S&P 500 returned 18.4% with dividends. In 2008 RIEF fell about 16%, while Medallion had one of its best years. Institutional Investor reported RIEF's annualized return since launch at about 8% (8.05%), against 9.6% for the S&P 500 over the same period.
Explanations reported at the time were that the public funds' models, built on months-long holding periods and historical patterns, could not adapt to the price swings of the pandemic and the speed of the rebound after the March 2020 crash; Renaissance itself declined to comment to Institutional Investor. The broader point is that being Renaissance does not make a fund extraordinary. What made Medallion extraordinary was a particular kind of edge, short-term and capacity-limited, that the outside funds do not have.
What it means for everyone else
- A real edge is usually small and fragile. Even Renaissance's best signals were reportedly right only a little more than half the time. Their value came from scale, leverage, low costs and discipline over millions of trades, not from being right about any one stock.
- Edges have capacity limits. The more money uses a pattern, the less it pays. That is one reason most active funds trail their index: large funds cannot exploit small, short-lived opportunities.
- The record you can buy is not the one you read about. Investors who wanted "Simons" exposure got RIEF, not Medallion. Ask which product a famous record actually belongs to.
- Patterns in data are easy to find and hard to trust. Renaissance tested ideas across decades and many markets and still threw most of them away. A rule that worked in one backtest is far weaker evidence; see Process over outcome: judging investment decisions under luck.
How Finabase helps
Finabase cannot show you Medallion's trades, but it lets you see why simple signals are harder to use than they look. Backtest runs a rule over two years of real prices so you can see how often it wins and how much costs and timing matter. The scanner lists stocks meeting statistical conditions each day, such as unusual volume or an extreme RSI, which is the raw material quantitative strategies start from. A long-run S&P 500 chart shows the 2000, 2008 and 2020 episodes in which Medallion reportedly did best.
Frequently asked questions
Can I invest in the Medallion Fund?
No. Medallion has been closed to outside investors since the early 2000s and is owned by current and former Renaissance employees. Renaissance's public funds, which outsiders could buy, use different, longer-term strategies and have had very different results.
What returns did the Medallion Fund make?
According to Zuckerman's 2019 book, Medallion averaged about 66% a year before fees and about 39% after fees from 1988 to 2018. These are averages of yearly results from private records, not audited public figures.
How did Jim Simons make money?
His firm used mathematical models to find many small, repeatable patterns in prices, traded them automatically with leverage and hedging, and kept the main fund small enough that the patterns kept working. Simons managed the scientists and the firm rather than picking trades himself.
Why did Renaissance's public funds lose money in 2020?
Reported explanations are that their models could not adapt to the pandemic's unusual price patterns and the speed of the recovery after the March 2020 crash. The public funds hold positions for six months to a year by one investor's account and could not adapt as quickly as Medallion, whose positions last minutes to days.
Summary
Jim Simons turned a mathematician's approach into the most successful fund on record. Medallion's reported results, about 66% a year before fees and 39% after from 1988 to 2018, came from many small, short-term statistical edges, traded with leverage and kept to a size the edges could bear. That last condition is why the fund is closed and why the firm's public funds, built by the same people for far more money, have produced ordinary returns and a large loss in 2020. For other investors the useful lessons are about the size, fragility and capacity of any real edge, and about checking which product a famous record actually belongs to.
Check your understanding
4 questions on this lesson. Each answer comes with the reason, and nothing is scored.
Why is the Medallion Fund closed to outside investors?
- Its short-term strategies can only absorb a limited amount of money before trading costs eat the edge, so the firm caps its size and keeps it for employees
- Regulators forbid outsiders from buying hedge funds
- It lost money and outsiders withdrew
Show the answer
A. Its short-term strategies can only absorb a limited amount of money before trading costs eat the edge, so the firm caps its size and keeps it for employees Each trade moves prices a little. Short-lived patterns can be exploited only at a limited scale, so Renaissance pays out profits and keeps the fund small.
Medallion reportedly averaged about 39% a year after fees from 1988 to 2018. Why is "averaged" important?
- It means the figure is a compound annual growth rate
- It is an average of yearly results; the fund paid out profits rather than compounding them, so money did not grow at 39% a year in the usual sense
- It means the figure includes only the best years
Show the answer
B. It is an average of yearly results; the fund paid out profits rather than compounding them, so money did not grow at 39% a year in the usual sense An arithmetic average of yearly returns is not the same as a compound rate, and a fund that distributes its profits does not compound its capital.
In 2020 Medallion gained 76% while Renaissance's public funds lost roughly 20% to 34%. What is the best explanation?
- Different people ran the public funds
- The public funds hold positions for months to manage far more money, so they rely on different, slower signals than Medallion
- The public funds were index funds
Show the answer
B. The public funds hold positions for months to manage far more money, so they rely on different, slower signals than Medallion Same firm, different product. Longer holding periods let a fund manage tens of billions, but the signals available at that horizon are weaker and more crowded.
Renaissance's best signals were reportedly right only slightly more than half the time. How could the fund still earn so much?
- It made a very large number of small trades, each with a small positive expected gain, and used leverage and hedging to turn small edges into large returns
- It made a few large bets on individual stocks
- It only traded during market crashes
Show the answer
A. It made a very large number of small trades, each with a small positive expected gain, and used leverage and hedging to turn small edges into large returns A small edge repeated many times, at low cost and with hedged leverage, adds up. The same logic explains why an edge that cannot be repeated often is hard to use.
Sources and further reading
- Zuckerman — The Man Who Solved the Market: How Jim Simons Launched the Quant Revolution, 2019 (Medallion returns 1988–2018, fees, history)
- Institutional Investor — Renaissance’s Medallion Fund surged 76% in 2020 but funds open to outsiders tanked, 13 January 2021
- Bloomberg — Renaissance hedge funds see red in 2020 while market turns green (investor letters), 2020–2021
- Chern and Simons — Characteristic forms and geometric invariants, Annals of Mathematics 99(1), 1974
- Simons Foundation — James H. Simons (1938–2024), biography and obituary, May 2024
- S&P Dow Jones Indices — S&P 500 annual total returns (2008, 2020)
- Finabase chart data — ^GSPC quarterly closes, 1984 to 2026
